Questions of coherence (COH?)-

Similar Questions

• Question of coherence 3.5 Beta

  I just started to look at version 3.5 beta and noticed some very interesting improvements in the release notes. I am especially interested to learn more (and testing!) improvements in storage NIO allowing more than 2 GB be used with 64-bit JVM:s. is implemented as an improved version of the current NIO memory manager or is it a new memory manager? I ask because I would like to know were I can fing documentation on how to use / configure it... Pointers to documentation is highly appreciated!
  
  I would also like to know if the new solution can cooperate with overflow even card in the case were the objects are not the same size (ie you must specify the HWM, in bytes and not in the number of objects)? It was not possible with the old NIO solution, but I'm keeping my fingers crossed that it's resolved with the new... If this is not fixed would it be possible/recommended to rely on virtual memory operating systems of handling in (at least for us very rare) situation were a memory overflow situation that otherwise would occur? Or is the cause of writing code in a way that blocks of memory allocated are all assigned both to a size maximum there paging happen immediately? Or maybe the memory allocated through IAN must be 'physical' memory?
  
  Best regards
  Magnus
  
  Published by: MagnusE on April 15, 2009 10:42

  Magnus,

  First of all, I apologize for the confusion between a question a little. The 'high-units' in the attached schema element would serve as the upper limit for the map full of support, not only a single partition. AAMOF, that's why we had to introduce the notion of "unifying factor" - for specifying capabilities beyond 2 GB limit without breaking backward compatibility (which would happen if we simply changed the type of data returned on the method getHighUnits of int and long).

  Now, answer questions from Robert:

  [1. you will have a single \[multiplexing\] listener for support partitioned plan

  2. There are indeed a number of new methods on the [BackingMapManagerContext | http://download.oracle.com/otn_hosted_doc/coherence/342/com/tangosol/net/BackingMapManagerContext.html] you will like:
  -getKeyPartition int (object oKey);
  -The getPartitionKeys (String sCacheName, int nPartition) value;
  -Card getBackingMap (String sCacheName);

  Kind regards
  Gene

• read plsqlchallenge question of coherence

  Hi Experts,

  Today, I saw a good question in reading consistency plsqlchallange.com I want to share with you. The interesting thing is, I thought that the cursor may receive prior to extraction. However, he said that it can be extracted when cursor connection occur. Is she not devoid of meaning? Because there is also a keyword called FETCH, if the cursor to retrieve data, so why we use the FETCH keyword too?

  My second question is, as far as I know in FOR extraction of cursor loop 100 lines for each iteration. As opposed to simple loop. I wonder if Oracle does not have the functionality of reading consistency, will be the display still loop the same result because it gets 100 lines per iteration?

  I create and populate a table as follows:
  CREATE TABLE plch_employees
  (
  employee_id INTEGER
  , last_name VARCHAR2 (100)
  Number of salary
  )
  /

  BEGIN
  INSERT INTO plch_employees
  VALUES (100, 'Smith', 100000);

  INSERT INTO plch_employees
  VALUES (200, "Jones", 1000000);

  COMMIT;
  END;
  /
  Blocks leading to the following two lines of text to display after running?
  Smith
  Jones

  BEGIN

  FOR SheikYerbouti IN (SELECT *)

  OF plch_employees

  ORDER BY employee_id)

  LOOP

  DELETE FROM plch_employees;

  Dbms_output.put_line (emp_rec.last_name);

  END LOOP;

  END;

  DECLARE
  CURSOR emps_cur
  IS
  SELECT * from plch_employees
  ORDER BY employe_id;

  SheikYerbouti emps_cur % ROWTYPE;
  BEGIN
  OPEN emps_cur.
  DELETE FROM plch_employees;

  LOOP
  GET emps_cur INTO SheikYerbouti.
  EXIT WHEN emps_cur % NOTFOUND;

  Dbms_output.put_line (emp_rec.last_name);
  END LOOP;
  CLOSE Emps_cur;
  END;

  Concerning

  Charlie

  Robert Angel wrote:

  Open cursor creates your snapshot that is consistent and is used by any code from there.

  Common misperseption opening cursor creates a snapshot. Is not! Opening slider simply records current SNA which becomes the reference point of read-consistency level statement. When the extraction is issued it read table and check the SNA of each block. If block YVERT > open cursor Oracle YVERT realize block changed since the cursor was opened, and will go to CANCEL to the previous status of block. He will check that previous state of block SNA and if it is new > open cursor YVERT repeats the process. Finally he will find either block consistent state in UNDO or trigger of"snapshot too old".

  SY.

• Acronyms of virtual Security Appliances

  It is more a question of coherence, rather than starting a new thread.

  But the acronym/abbreviation that will be used for the Web Cisco Security Appliance.  I saw vWSA and WSAv.  I would like just a bit of clarity, for reviewing products with my management.

  According to the official Cisco Data Sheet for Web Security Appliance below:

  http://www.Cisco.com/c/en/us/products/collateral/security/content-Securi...

  It's using WSAv, then maybe go with it.

• 12.2.1 migration - new generics introduce compilation errors

  I am currently moving projects consistent with 12.2.1 (since 12.1.2.x) and the code compiles more.

  I see that some types of coherence have been changed to generics and this introduces errors or type checking type mismatches.

  The Live event interceptor next used to work:

  Public MustInherit class MyInterceptor implements {EventInterceptor < EntryEvent >

  {} public void onEvent (EntryEvent evt)

  {for (be BinaryEntry: {evt.getEntrySet ())}

  . . .

  }

  }

  }

  Now, he seems to have will be replaced by:

  / public abstract class MyInterceptor implements EventInterceptor < EntryEvent <? > > {}

  {} public void onEvent (EntryEvent evt)

  {for (be BinaryEntry: ((Set < BinaryEntry <?,? >>)) {evt.getEntrySet ())}

  . . .

  }

  }

  }

  Any more elegant way to adjust this code?

  This is documented somewhere in the coherence 12.2.1 release notes?

  Unfortunately this is a problem with the Java compiler, more precisely how the compiler infers and 'excessively' clears type information in a type, including those of inherited types (classes / interfaces) in the absence of soft types.  That is to say: How do you treat the erasure.

  Let us look at the part of the EntryEvent interface as defined by the coherence that is the origin of errors you see.

  EntryEvent /public interface extends Event {}

  Public game<>> getEntrySet();

  }

  When the parameters of type are left out a declaration of EntryEvent, which is natural when you're upgrading the source code to consistency 12.2.1 these settings did not exist before the 12.2.1, consistency the Java compiler guess essentially that:

  1. Your type of EntryEvent is using type erasure, and
  2. All types defined using parameters of type (everywhere) should be deleted.

  Therefore, a statement like 'e EntryEvent'; becomes effectively;

  EntryEvent/public interface extends Event {}

  public Set getEntrySet();

  }

  as you can imagine, is not what is expected.

  According to the team of Java and Java Language Specification, this is how it is supposed to work.  However as we have shown, not only in the above example, but in the examples of different basis of consistency / breeding, current "interpretation" of the specification of the Java language seems a bit harsh.  At worst we think that only the types which have been left out are deleted, certainly not all the types and certainly no inherited or supertypes.   for example: at worst the infers of EventEntry type should look like this:

  EntryEvent/public interface extends Event {}

  public Set getEntrySet();

  }

  Unfortunately, it is a little worse than that, especially when you consider that the great event interface defines the following method (where T extends Enum)

  public T getType();

  Unfortunately, when all soft types are deleted, what do you think of the type t becomes effectively?   If you guessed object, you're right.   But you would also correct (according to me) If you have accessed only one Enum!   So if your source using this method, you have now received warnings of Cast of the class or exceptions when you try to use the return as an Enum value

  The solution, as you have discovered, is to avoid the type erasure, or tell the compiler to not erase the types.  By changing the declaration of EntryEvent to EntryEvent or EntryEvent the compiler will prevent erasure of information essentially making and deduct all the types as object or type Super correct, for example: Enum instead of the object.

  We talked at length with the team of Java on this subject, including raise a fault against the OpenJDK (where such a fix should be applied).  Unfortunately the fix can break a wider range of the code in the Java community, and so we are unfortunate to have to walk around the anomaly of Java or adding (to avoid erasing type) or add casts.

  Technically, it is not a question of coherence, but in this case the corner, we hit with the consistent input event.

  The bottom line seems to be: If you want to use generic drugs, you need to do it everywhere, especially to be safe and avoid possible compiler problems, who are still kicking on (since Java 5!).

  Sorry I don't have a better answer.

  -Brian

  Architect | The Oracle coherence

  PS: I'll take a look at the release notes.   He should be here... because I wrote the note of specific version on this issue.

• Oracle weblogic coherance Questions

  Helllo Team,
  
  Could you please help me to understand about consistency.
  
  1: what is the oracle weblogic server consistency, how it useful in real time scenarios.
  (2) that I saw in the documents of the oracle, he was in the memory grid, is it's keep business logic, in this case in weblogic cluster as we use even in memory replication, how this different between cluster weblogic.
  
  (4) is coherance is a database.
  
  Kindly help me questions to understand on oracle weblogic server consistency .is there any simple document above or liniks also very well.
  
  Thank you
  Reddy

  Maybe this (http://middlewaremagic.com/weblogic/?p=8351) can help you on the way.

  This (http://docs.oracle.com/cd/E24290_01/coh.371/e22620/start.htm#CIHHAAIH) provides a good summary of what coherence * Web.
  «Coherence * Web is a HTTP session management module dedicated to the management of session state in a clustered environment.» Built on top of the coherence of the Oracle (consistency), coherence * Web:
  -brings data grid of data consistency, scalability, availability, reliability and performance for the session in memory management and storage.
  -can set fine-grained and session attribute carried through pluggable policies (see "Session and Session attribute framing).
  -can be deployed on multiple servers to consumer applications such as Oracle GlassFish Server, Oracle WebLogic Server, IBM WebSphere, Tomcat and so on (see "Supported Web containers").
  -can be deployed on several Portal containers, including Oracle WebLogic Portal (see Chapter 10, "Using coherence * Web with WebLogic Portal").
  -allows storage of session data outside the Java EE application server, freeing heap for the application server space and allow the restart of the server without loss of session data (see "Deployment Topologies").
  -allows the sharing of session and management across different Web applications, domains and servers in heterogeneous applications (see "Session and Session attribute) scope.
  -can be used in models of advanced session (that is, traditional, monolithic and split Session) that define how session state is serialized or deserialized in the cluster (see templates "Session"). »

• Questions about an average of response spectrum and frequency of feeding mode.

  Hello

  I have a few questions about an average of mode. When I generate a sinusoidal signal from one output to two input channels channel to see if my DAQ card works well and vector averaged in the power spectrum for DFT, the amplitudes was different from the previous one of the amplitude, which was supposed to be 1 v peak. They range from 0.5 v to 0.6 v peak. When the calculation of the average model is RMS, the amplitudes were close to 1. I wonder what are the fomulas of RMS and average vector. Does that mean that I could not accurate if I use an average of vector? In a time of frequency response, why I coherences of difference and the amplitudes using the vector and the mean quadratic value?

  Thank you

  Ningyu

  rico1985,

  The differences in modes of generation are as they sound: 1 sample output only a sample writing, N samples will be released however many samples configure you for each entry, and the continuous samples released samples continuously until a specified user action happens (you press the stop button or a logic that you created gets fulfulled). The range of Signal output allows you to set a ceiling high and low level of your output signal and it only affects the quality keeping in this beach. Timing to set a deadline for the time between the acquisition of the sample. If a new sample becomes unavailable before the timeout setting, you will get an error. This is useful for looking at a network, because if the network goes down and you stop getting data from a machine and then you would like to know about it. I point you to those videos that are short tutorials on how to make the most of these actions in SignalExpress.The SignalExpress 3.0 Help file is also your first point of contact for all your questions on getting started. These two resources should get you up and running in SignalExpress in no time. (either by the way all your questions answered using these resources) Bravo!

• Problems upgrading to 12.3.1 coherence incubator

  I am new to the use of consistency. When you try to upgrade v 3.7 and the incubator version correspondent am faced with the question at the start of the server. Indications as to why this would will help me narrow down where I should look.

  Exception in thread "Thread-5" java.lang.IllegalArgumentException: ensureCache cannot find a schema for cache coherence.patterns.processing.taskprocessordefinitions

  at com.tangosol.net.ExtensibleConfigurableCacheFactory.ensureCache(ExtensibleConfigurableCacheFactory.java:220)

  at com.oracle.coherence.patterns.processing.internal.task.DefaultTaskProcessorDefinitionManager.onDependenciesSatisfied(DefaultTaskProcessorDefinitionManager.java:261)

  at com.oracle.coherence.patterns.processing.internal.ProcessingPattern.start(ProcessingPattern.java:170)

  at com.oracle.coherence.patterns.processing.internal.ProcessingPattern.ensureInfrastructureStarted(ProcessingPattern.java:123)

  to com.oracle.coherence.patterns.processing.internal.LifecycleInterceptor$ 1.run(LifecycleInterceptor.java:59)

  at java.lang.Thread.run(Thread.java:724)

  He realized that the cache of the incubator configs are not load. Can't understand where reference them for charging. The next difference are older newspapers:

  2015-12-24 11:49:59.446/9.086 Oracle coherence GE 3.7.1.0 < Info > (thread = main Member, = n/a): loaded cache configuration of "jar:file:/D:/Workspace/.metadata/.plugins/org.eclipse.wst.server.core/tmp0/wtpwebapps/pas.web/WEB-INF/lib/coherence.patterns.processing-1.4.2.jar!" " /processingpattern-coherence-cache - config.xml.

  2015-12-24 11:49:59.478/9.118 Oracle coherence GE 3.7.1.0 < Info > (thread = main Member, = n/a): loaded cache configuration of "jar:file:/D:/Workspace/.metadata/.plugins/org.eclipse.wst.server.core/tmp0/wtpwebapps/pas.web/WEB-INF/lib/coherence.common-2.1.1.jar!" / " common-coherence-cache - config.xml"

  I use the following versions:

  < dependency >

  com.Oracle.coherence < groupId > < / groupId >

  consistency of < artifactId > < / artifactId >

  12.1.3 - 0-0 < version > < / version >

  < / dependence >

  < dependency >

  com.Oracle.coherence.incubator < groupId > < / groupId >

  < artifactId > coherence-processingpattern < / artifactId >

  < version > 12.3.1 < / version >

  < / dependence >

  Problem has been resolved. His changes to the appearance of extensibility. The xml of substitution of coherence in the old version is obsolete and is no longer used. Added the following namespaces for the thing to work.

  xmlns:element = "class://com.oracle.coherence.common.namespace.preprocessing.XmlPreprocessingNamespaceHandler".

  item: introduction-cache-config = "coherence-cache-config. XML">

• Announcing the coherence 3.7.1.10 for Java (Patch 10)

  Consistency Java 3.7.1.10 (Patch 10) aired in Orion (Metalink) as patch 17342587 .
  
  The patch contains the following hotfixes:
  COH-10313: fixed a problem of comparator custom service proxy load balancing.
  COH-10271: improve the sending of the DiagnosticPackets as they are sent to the announced and favorite ports.
  COH-10270: fixed an issue with the elastic data in which an invalid binary exception is reported during an operation which directly or indirectly called keySet.
  COH-10219: fixed an issue where the full value of-31 is not read correctly by the serializer POF.
  COH-10171: hardened transactional processing cache of javax.transaction.xa.Xid to allow implementations not serializable.
  COH-10161: Add a new configuration of session in coherence * Web, which maintains a local instance of a session in addition to hot flashes the session for the distributed cache.
  COH-10124: fixed an issue which can lead to false warnings of WKA list incompatible when using host names in the list of the WKA.
  COH-10119: fixed a problem of serialization infiltrate.
  COH-10115: fixed a problem causing the NamedCache iterators return results in duplicate to the clients to extend Standard or Enterprise edition clusters.
  COH-10078: fixed an issue where the RWBM.flush () could return before all the data is flushed.
  COH-10063: fixed an issue that caused the calculator unit configured to be ignored by the ReadWriteBackingMap.
  COH-10054: fixed an issue causing blocking between the start of the cluster and the thread starts a service.
  COH-10033: fixed an issue where the listeners on replicated Cache could receive card events.
  COH-9967: fixed an issue that could cause the memory of the customer take if the distribution of the "stuck" partition
  COH-9934: fixed a problem with the filter run producing the overload of high memory due to the useless JMX statistics.
  COH-9902: Fixed problem of intiailation potential session in WebLogic during side by
  COH-9786: added support of coherence * Tomcat 7.x application servers Web.
  COH-9690: hardened treatment of disabling cache calls
  COH-6587: hardened treatment of configuration errors of cache to guard against NPES and UOE to offer a more stable service.
  
  Fix full list of all Java 3.7.1 consistency fixes are available in both the patch readme file and in Note: 1405110.1

  This has now been replaced by the 3.7.1.11 output.

• PeopleTools 8,53 Linux Version PIA: JDK 7 questions

  People,

  Hello. I install PeopleTools 8.53 (Linux Version) with the generic WebLogic11gR1 and coherence (name of file V29856-01) which requires JDK 7 Linux x 86-64 (V35017-01 file name). All files are downloaded from http://www.edelivery.oracle.com.

  Before installing WebLogic 11 GR 1 material, install the JDK 7! After having extracted v35017 - 01.zip under Linux, it turns into the file jdk-7u9-linux - x 64 .rpm.

  During the installation of the file, I get the message as below:

  [root@localhost Java7_JDK_Linux_64bit]# rpm -ivh jdk-7u9-linux-x64.rpm
  
   Preparing... ########################################### [100%]
   1:jdk ########################################### [100%]
   Unpacking JAR files...
   rt.jar...
   Error: Could not open input file: /usr/java/jdk1.7.0_09/jre/lib/rt.pack
   jsse.jar...
   Error: Could not open input file: /usr/java/jdk1.7.0_09/jre/lib/jsse.pack
   charsets.jar...
   Error: Could not open input file: /usr/java/jdk1.7.0_09/jre/lib/charsets.pack
   tools.jar...
   Error: Could not open input file: /usr/java/jdk1.7.0_09/lib/tools.pack
   localedata.jar...
   Error: Could not open input file: /usr/java/jdk1.7.0_09/jre/lib/ext/localedata.pack
  
   [root@localhost Java7_JDK_Linux_64bit]# 
  
  
   Some folks told me the above message is a warning and can be ignored. However, I have deleted the default installation directory /usr/java and install it again and get the message below:
  
   [root@localhost Java7_JDK_Linux_64bit]# ls
   jdk-7u9-linux-x64.rpm V35017-01.zip
  
   [root@localhost Java7_JDK_Linux_64bit]# rpm -ivh jdk-7u9-linux-x64.rpm
   Preparing... ########################################### [100%]
   package jdk-1.7.0_09-fcs is already installed
   [root@localhost Java7_JDK_Linux_64bit]# 
  
   [root@localhost Java7_JDK_Linux_64bit]# rpm -ivh --prefix=/JDK7 jdk-7u9-linux-x64.rpm
   Preparing... ########################################### [100%]
   package jdk-1.7.0_09-fcs is already installed
   [root@localhost Java7_JDK_Linux_64bit]#
  
  

  
  The installation by default is/usr/Java directory, but the package of jdk - 1.7.0_09 - fcs is not in/usr/java. I am looking for the entire file system, but the packagejdk - 1.7.0_09 - fcs is not in other directories.

  My questions are:

  First of all, the package of jdk - 1.7.0_09 - fcs is not any directory. Why the message says "package jdk - 1.7.0_09 - fcs is already installed?

  Second, I try to install it in another location /JDK7 and get the same message as above. What's wrong?

  Thirdly, do I need to remove the directory/usr and then reinstall it?

  
  Thank you.

  Code formatting of Message was edited by: NicolasGasparotto

  Do not delete / usr, or you will be reinstalling your operating system.
  because you installed with rpm, you should have removed with rpm. remove the files installed by RPM does not remove the package from the rpm database.
  try to make

  rpm - Uvh - force jdk-7u9-linux - x 64 .rpm

  to reinstall over the existing package

  RPM man

  gives you all the documents and there are of countless pieces of information available online. It seems that you should read up on the basics of Linux Administration. Administration/install PeopleSoft requires skills in operating systems, without them, that you will meet potential problems in all directions.

• When the memory coherence is full, what consistency do?

  Dear Expert:
  
  I have a problem in customer, they ask:
  When the memory coherence is full, what consistency do?
  He takes care to write data or the entry on the disk?
  If it can be supported, how can I config?
  
  
  I read some documents, but I find things like this:
  < local plan >
  LRU <-eviction strategy > < / eviction strategy >
  < high-units > 1000 < / high units >
  < expiry delay > 1 h < / timeout >
  < / local plan >
  
  but I can not find somethings on the coherence jvm memory usage?
  
  
  Thanks in advance.

  Hello

  There is no magic solution to stop short of memory coherence. A cluster of coherence is a limited resource and has a finite size and you need to do some sort of exercise as you would with something like a database capacity planning system. You wouldn't continue to put data in a database without him running out of resources, and the same is true with consistency.

  Yes, there are things like expiry which allows you to limit the size of the caches, but like the previous poster said, when the limit is reached data will be evicted from memory. Unless you have some form of persistent storage for these data using something like a dumps, while the data is lost.

  You can set the overflow to disk so that when a size the cache limit is reached as described here http://download.oracle.com/docs/cd/E15357_01/coh.360/e15723/cache_examples.htm#BACCHCIA

  The problem with the ousting of size based and exceeded the size in function, is that the limit is per cache per JVM. This means that for a system with several caches and services that you usually end up expulsion or overflow to disk for a cache that reached its limit when there still may be a lot of memory left that you need to allow for each cache being full in your design.

  I saw people in the past are trying to find different ways of ousting of data based on the used JVM heap, but they never seem to be satisfactory. If you've seen a chart of the JVM heap usage you will see that it is usually a model of Sawtooth constantly up and down. A JVM could hit 90% use but still have a lot of real space because he had not recently made a GC. If you have a schema data eveicted when the reached JVM say 90% you could start to expel the data too early. Also, it's probably enough data in a cluster push possile faster than data can be evicted if you might still run out of memory. A cluster of coherence consists of multiple JVM processes and if you get skewed the disaggregation of data or you caching of objects of different sizes, it is quite possible to get Java virtual machines in the cluster running off lot well before others. The project I'm working on that, we had a situation where we had a few nodes in the cluster using more than 300 MB of heap more than the other nodes in the cluster - is quite a big difference when the total heap was only 2.5 GB.

  As I said the best thing to do is good capacity planning. Work out how much you need to organize and work on the size of your cluster based on that data. You probably still have some sort of eviction of the data well. The system on which I work at eviction of the moment according to some data and custom eviction of other data based on business rules. Custom eviction is juats another process that verifies the data and expels the relevant stuff.

  JK

• Can limit you the size of lot writeback for a defined number of coherence 3.5?

  I am running on Windows 3.5.3p9 consistency.
  
  The cache store is configured to use the schema to write back via the tag reading-writing-support-map-plan.
  
  The batch is activated with a write delay of 5s.
  
  My understanding is that everything that has been newly inserted into the cache more than five seconds ago becomes basically eligible for storage.
  
  Our application through a bit of a cycle of peaks and valleys. Sometimes very little data is inserted at once and is sometimes a lot. This results in quite different lot sizes and size jackpots cause problems on our database from time to time.
  
  I can decrease the delay-write bits to 1 in the hope that this will decrease in turn in the size of the lots, but is there a way to set a specific number, for example, I want never to write 20 in a batch?

  Hello

  You can just break down that big jackpot in smaller batches (DB transactions) yourself, and you can also decide that you do not want to write more for the moment.

  If you get a consistency exception will try again everything that is in the mapping of parameter passed to the storeAll and eraseAll parameter collection. But it doesn't have to be complete, it is expected that you delete entries / items that you persisted successfully.

  In this way, you can control the rate of write yourself. In addition, given that the lazywriter thread does not block the processing of the event, so you're kind of danger waiting in these methods if you want a little space resources transactions without return storeAll.

  To answer your question:

  You can either

  -make a physical transaction of 20 elements, remove 20 items on the map, then possibly wait and then continue with more elements of the card as long as it takes place (this gives you the ability to control the rate of transactions).

  -Send a physical transaction of more than 20 items to the database, remove the 20 items on the map and then thrown a dummy (in which case coherence-requeues the rest... Take care, after that they are considered as freshly changed entries).

  Best regards

  Robert

• Coherence license for Hyperthreaded cores?

  Given that the consistency of our days is allowed only by code (and not by CPU as possible at the time Tangosol) an interesting question is what the hearts (on Intel HW) threaded hyperthreading are managed?
  
  Cores HT adds only a fraction of the capacity of the real cores so that if they are allowed as a normal kernel function HT must be disabled in order to get good value for money...
  
  / Magnus

  Tangosol days, assuming that I recall, hyperthreading has been addressed by detecting the CPU and to halve the number of reported if core hyper-threaded processor.

  Oracle has a system of well-defined, license that I refer you to:
  http://www.Oracle.com/us/corporate/contracts/license-service-agreement/index.html

  It appears suddenly look that they measure the real cores, no concurrent threads (for example, hyperthreading).

  Peace,

  Cameron Purdy | The Oracle coherence
  http://coherence.Oracle.com/

• Warning of global coherence.

  I created an association of security with four tables involved. Everything was going well and I ran reports that RPD. Now, I added a table over in the physical layer to create an another SA involving only two tables. First it was a star schema in the physical layer, but after I did this painting he became a snowflake schema. But in MDB as the second SA covers only two tables I have not met this snowflake in MDB. HIS first has 4 tables which is a star schema and ITS second in question only two tables. Now after I am dome with everything and pulled the new presentation layer and checked a global coherence, I get a warning that says "39011 keys 'Topic Area2'. table.table_key #2 in the logical table" topic Area2. "table" is a superset or subset of another key in this table. Any redundant table key must be removed. What should I do about it.

  Hello
  Double-click the source table, keys tab remove the extra key "#" then check the consistency.

  Thank you
  saichand.v

• Define the new service of tangosol-coherence - override.xml

  Hi all
  I am new to consistency. Have a problem to define the new service.
  
  I am trying to create a new service replied to a few caches and intend to run this service only on some nodes in the cluster.
  
  So I created the new definition of service in tangosol-coherence - override.xml by copying the definition for ReplicatedCache (id = "1"). I changed the ID number to + 101 + and changed the service type for "newReplicatedCache."
  Then in my cache - config.Xml. I put the tag < service name > "myService" to "newReplicatedCache".
  
  It starts and runs ok, and Jconsole MBean I see there is a "newReplicatedCache" with "myService" running, which is great.
  
  Now, the problem comes, tangosol-coherence - override.xml, <>lease-granularity for "ReplicatedCache" is set to Member.
  And I need put < lease-granularity > to 'newReplicatedCache' to wire.
  
  I did and seems consistency is without taking account of this parameter and always using members to 'newReplicatedCache '.
  
  Then I removed the definition for 'newReplicatedCache' in tangosol-coherence - override.xml, and consistency is always be able to run, and there always create a new service "newReplicatedCache". I guess this is just to get the name of cache-config. XML and then use "ReplicatedCache" default service definition in column a new service.
  
  So my question is how to define new services?
  
  I checked the documentation for the serial number, it does not explain how to create a new service. I understand that this is a very simple matter of installation, will be more than happy to read the documentation if you can direct me to that useful :)

  Hello

  To achieve what you want, delete ... your file tangosol-coherence - override.xml and change your schema definition reproduced in you cache configuration to:

  
    myNewCache-scheme
    myNewReplicatedCache
  
     thread
  
    
      
      
    
  
  

  Andy